Private Learning of Littlestone Classes, Revisited
This work addresses privacy-preserving learning for structured classes, offering significant improvements in efficiency for applications requiring differential privacy, though it is incremental in building on existing techniques.
The paper tackles the problem of online and PAC learning of Littlestone classes under approximate differential privacy constraints, achieving a mistake bound of $ ilde{O}(d^{9.5}\cdot \log(T))$ in the realizable case, which is a doubly-exponential improvement over prior work and nearly matches the lower bound.
We consider online and PAC learning of Littlestone classes subject to the constraint of approximate differential privacy. Our main result is a private learner to online-learn a Littlestone class with a mistake bound of $\tilde{O}(d^{9.5}\cdot \log(T))$ in the realizable case, where $d$ denotes the Littlestone dimension and $T$ the time horizon. This is a doubly-exponential improvement over the state-of-the-art [GL'21] and comes polynomially close to the lower bound for this task. The advancement is made possible by a couple of ingredients. The first is a clean and refined interpretation of the ``irreducibility'' technique from the state-of-the-art private PAC-learner for Littlestone classes [GGKM'21]. Our new perspective also allows us to improve the PAC-learner of [GGKM'21] and give a sample complexity upper bound of $\widetilde{O}(\frac{d^5 \log(1/δβ)}{\varepsilon α})$ where $α$ and $β$ denote the accuracy and confidence of the PAC learner, respectively. This improves over [GGKM'21] by factors of $\frac{d}α$ and attains an optimal dependence on $α$. Our algorithm uses a private sparse selection algorithm to \emph{sample} from a pool of strongly input-dependent candidates. However, unlike most previous uses of sparse selection algorithms, where one only cares about the utility of output, our algorithm requires understanding and manipulating the actual distribution from which an output is drawn. In the proof, we use a sparse version of the Exponential Mechanism from [GKM'21] which behaves nicely under our framework and is amenable to a very easy utility proof.